Complementary but Different: The Politics of Green Industrial Policy and Carbon Pricing

Sweden, historically a global leader in carbon pricing, has recently made a significant shift in its climate policy towards green industrial policy. It has moved away from environmental taxation – reflected in reduced transport fuel tax rates and increased emissions from the transport sector – towards a state-driven energy policy centered on nuclear power. To support the planned construction of ten new nuclear reactors, the Swedish government has proposed loan guarantees and state loans of up to $40 billion (Persson, 2022). By lowering transport fuel tax rates while simultaneously offering state support for nuclear energy, Sweden is treating carbon pricing and green industrial policy as substitutes rather than complements. This policy brief challenges that approach, arguing that carbon pricing and green industrial policy should be seen as complementary climate policy instruments. However, their political economies differ significantly, making industrial policy more politically feasible. Yet, the two key challenges with green industrial policy are how to finance it and how to “pick winners” – choosing which technologies and companies to support. We use the recent bankruptcy of Swedish battery manufacturer Northvolt as a case study to illustrate these challenges.

Two Climate Policy Instruments

Over the past decade, climate policymaking has undergone a significant shift. Governments in Europe, USA, and China have increasingly adopted green industrial policy as a central strategy to reduce carbon emissions, moving beyond the traditional focus on carbon pricing.

Green industrial policy serves both environmental and economic purposes. It mobilizes government efforts toward decarbonization while fostering the development of zero-carbon technologies and domestic firms to drive employment, innovation, and growth in green sectors (Rodrik, 2014). Common policy tools include subsidies, loan guarantees and state loans. Sweden’s emphasis on loan guarantees and state loans to support nuclear energy is not unusual, as these are among the most widely used industrial policy instruments in rich countries (Juhasz, Lane, and Rodrik, 2023). A historical parallel can be found in France, which relied on state loans and loan guarantees in the 1970s and 1980s to support its nuclear energy expansion (Andersson and Finnegan, 2024).

Carbon pricing, on the other hand, focuses solely on emissions by putting a price on the negative externality of carbon emissions. By equalizing the private and social cost of releasing carbon, carbon pricing leaves it to the market – firms and households – to decide how to most effectively reduce emissions. While carbon pricing is a form of government intervention to correct a market failure, it is technology neutral, and the state is not actively steering the economy toward specific pathways of decarbonization in the same way that green industrial policy does.

Complements Rather Than Substitutes

At its core, decarbonization requires shifting the cost advantage from fossil fuels to the zero-carbon alternatives of wind, solar, and nuclear. The key factor is the relative price between the two energy categories. This shift can be achieved by increasing the cost of fossil fuels through carbon pricing or by lowering the cost of zero-carbon energy via industrial policy.

The cost of energy production – particularly electricity – can be divided into three main components: capital investment costs, operation and maintenance, and fuel costs (IEA, 2005; 2020). While operation and maintenance costs are typically a minor share of the (levelized) cost of electricity for all energy sources, there are large differences in investment and fuel costs between fossil fuels and zero-carbon alternatives. The cost of fossil fuels is sensitive to fuel prices, which often account for more than half (coal), to around 80 percent (natural gas), of total costs in most regions. Wind and solar, on the other hand, have zero fuel costs but are highly dependent on capital investment costs, making them particularly sensitive to interest rates. Nuclear energy, though requiring some fuel costs, is also predominantly capital-intensive (IEA, 2015).

Because of these differences in cost structure between the energy categories, carbon pricing and industrial policy work as complements rather than substitutes. Carbon pricing raises the variable fuel costs of fossil fuel-based energy, making it less competitive, while industrial policy can reduce the fixed capital costs of zero-carbon technologies, improving their affordability. A well-balanced climate strategy may employ both approaches to achieve decarbonization. A revenue-neutral model could even use carbon pricing revenues to fund industrial policy, balancing cost burdens and investment incentives.

Figure 1 illustrates how the two policy instruments of carbon pricing and industrial policy are complements when it comes to climate policy as they both shift the relative price in favor of zero-carbon energy sources.

Figure 1. Decarbonization and relative prices

Source: Authors’ illustration.

Differences in Their Political Economy

Despite their complementarity, the political economy of carbon pricing and green industrial policy differs significantly, making the latter more politically feasible.

First, carbon pricing and green industrial policy differ in how they distribute costs and benefits across time and geography. Carbon taxes impose short-term, localized, and visible costs on consumers and producers while generating long-term, globally dispersed benefits. Because the costs and benefits are unevenly distributed over time and space, the households that bear the costs are likely not the same as those that receive the benefits. In contrast, green industrial policy can create immediate and visible local benefits for households and businesses, while spreading the costs more broadly. These costs can be distributed nationally using general taxation, internationally through global climate funds, or shifted into the future via deficit spending.

Second, carbon pricing generates a first-mover disadvantage, as the implementing country will incur higher energy prices for producers and consumers and thus potential deindustrialization and unemployment as firms relocate to countries with less stringent climate policy and lower energy costs. Green industrial policy inverts this narrative by incentivizing low-carbon firms to relocate to countries offering substantial state support. As a result, the first country that adopts generous green subsidies will put political pressure on other countries to do the same for fear of job loss and diminished competitiveness. This dynamic has been in play over the last couple of years with European leaders fearing the impact on European competitiveness of the Inflation Reduction Act in the U.S. – the largest climate bill ever implemented in that country – and green industrial policy in China. In this sense, subsidies offer a first-mover advantage, encouraging early adoption.

Combined, these two important political economy factors make green industrial policy more politically feasible by increasing public and political support compared to carbon pricing.

The first-mover advantage of green industrial policy also has important implications for global climate policy. The advantage, coupled with increasing opportunity costs of non-adoption (loss of competitiveness), can result in an equilibrium where the largest economies, such as the U.S., China, and the EU, all adopt similar green industrial policies. The first country that adopts green industrial policy pressures other nations to follow suit, fearing job losses, diminished competitiveness, and market-share erosion, creating a domino effect that results in a global implicit carbon price. This outcome is an equilibrium since none of the “players”, observing the choices made by others, have an incentive to withdraw their state support and subsidies for the green sector.

In contrast, a globally imposed carbon price using taxes, such as through international agreements like the Paris Agreement, does not constitute an equilibrium. Countries under such an agreement continuously face incentives to defect by repealing their carbon taxes to gain competitive advantages and free-ride on the ambitions of others. To transform such an agreement into a stable equilibrium, there must be credible punishment mechanisms – such as border carbon adjustments that penalize imports from defecting countries – to reduce incentives for free-riding (Nordhaus, 2015). Yet, such a global agreement with credible punishments has remained elusive, reflecting the complexities of international cooperation.

Two Key Challenges

While politically more feasible compared to carbon pricing, governments face two key challenges with industrial policy: how to finance it and how to select the right technologies and companies to support. These challenges are not just theoretical – they have real-world consequences. The recent failure of the Swedish battery manufacturer Northvolt highlights the potential risks governments face when using industrial policy.

Founded in 2015, Northvolt aimed to supply batteries for electric vehicles and energy storage, positioning itself as Europe’s main competitor to dominant Chinese manufacturers. With a rapid expansion of factories, the company struggled with production delays, mounting losses, and an inability to secure additional capital investments, ultimately leading to its bankruptcy. The Swedish government has provided some economic support but was unwilling to match the kind of large-scale state subsidies that China provides to its battery industry (Ekström och Mikaelsson, 2024). Likely, the level of financial support required for Northvolt to compete globally would need to come from the EU level, rather than national funding alone (Milne et al., 2025). The Northvolt case emphasizes one of the main challenges for green industrial policy: financing. Unlike carbon pricing, which generates revenue, industrial policy requires substantial government funding. High fiscal costs may limit its feasibility outside of large economies like China, the U.S., and the EU.

Furthermore, even if Sweden had provided stronger financial support – similar to its proposed subsidies for nuclear energy – Northvolt may still have failed due to technological competition. Experts suggest that Chinese competitors will be reluctant to acquire Northvolt’s Swedish factory, as Chinese investors believe it’s not correctly constructed for battery manufacturing (Nordensson, 2025). This underscores the second risk of industrial policy: governments may invest in technologies and companies that ultimately fail to compete.

Conclusion

Carbon pricing and green industrial policy are complementary tools for climate mitigation, but their distinct political economies make industrial policy more politically feasible. However, with green industrial policy, governments are faced with the risks of “picking winners” and how to finance the policy.

Sweden faces these two risks with its nuclear energy strategy. For instance, the levelized cost of nuclear energy has risen over time (Bilicic and Scroggins, 2023). Today, nuclear is the most expensive source for new grid capacity, while wind and solar are the cheapest. By 2045, when Sweden’s planned ten new reactors are expected to be operational, renewables may be so cheap that nuclear power struggles to compete, leading to financial losses and high electricity prices (SVT, 2024). In this sense, Sweden’s focus on a single zero-carbon technology may turn out to be a costly mistake.

Sweden should use green industrial policy as a complement to, rather than a substitute for, its previous carbon pricing strategy. Furthermore, to reduce the risks of not picking the “winners,” it should diversify its support across multiple zero-carbon technologies – including electric vehicles, battery manufacturing, solar, and wind – rather than focusing narrowly on nuclear power.

References

• Andersson, J. and Finnegan, J. (2024). Industrial Policy and Decarbonization: The Case of Nuclear Energy in France. Working Paper.
• Bilicic, G. and Scroggins, S. (2023). 2023 Levelized Cost of Energy+. Lazard.
• Ekström, J., and Mikaelsson, C. (2024). Därför nobbar regeringen Northvolt. Svenska Dagbladet. October 6, 2024.
• IEA. (2005). Projected Costs of Generating Electricity: 2005 Edition. International Energy Agency. Paris.
• IEA. (2015). Projected Costs of Generating Electricity: 2015 Edition. International Energy Agency. Paris.
• IEA. (2020). Projected Costs of Generating Electricity: 2020 Edition. International Energy Agency. Paris.
• Juhazc, R., Lane, N., and Rodrik, D. (2023) The New Economics of Industrial Policy. Working Paper 31538, National Bureau of Economic Research.
• Milne, R., Johnston, I. and Bounds, A. (2025). Boss of bankrupt Northvolt urges Europe to invest in homegrown battery sector. Financial Times. March 13, 2025.
• Nordensson, B. (2025). Expert: Inga utsikter driva Northvolt vidare. Svenska Dagbladet. March 12, 2025.
• Nordhaus, W. (2015). Climate Clubs: Overcoming Free-riding in International Climate Policy. American Economic Review, 105(4), 1339–1370
• Persson, I. (2022). Allt du behöver veta om ’Tidöavtalet. SVT Nyheter. 14 October, 2022.
• Rodrik, D. (2014). Green Industrial Policy. Oxford review of economic policy 30 (3):469-491.
• SVT Nyheter. (2024). Kärnkraften kan bli nära dubbelt så dyr som regeringen trott. SVT Nyheter. January 25, 2024.

Disclaimer: Opinions expressed in policy briefs and other publications are those of the authors; they do not necessarily reflect those of the FREE Network and its research institutes.